PaliGemma 2: A Family of Versatile VLMs for Transfer
This work provides versatile base models for researchers and practitioners to improve performance on diverse vision-language tasks, though it is incremental as an upgrade to PaliGemma.
The authors introduced PaliGemma 2, an upgraded family of vision-language models based on Gemma 2, trained at multiple resolutions to enable broad transfer via fine-tuning, achieving state-of-the-art results on tasks like OCR, captioning, and radiography report generation.
PaliGemma 2 is an upgrade of the PaliGemma open Vision-Language Model (VLM) based on the Gemma 2 family of language models. We combine the SigLIP-So400m vision encoder that was also used by PaliGemma with the whole range of Gemma 2 models, from the 2B one all the way up to the 27B model. We train these models at three resolutions (224px, 448px, and 896px) in multiple stages to equip them with broad knowledge for transfer via fine-tuning. The resulting family of base models covering different model sizes and resolutions allows us to investigate factors impacting transfer performance (such as learning rate) and to analyze the interplay between the type of task, model size, and resolution. We further increase the number and breadth of transfer tasks beyond the scope of PaliGemma including different OCR-related tasks such as table structure recognition, molecular structure recognition, music score recognition, as well as long fine-grained captioning and radiography report generation, on which PaliGemma 2 obtains state-of-the-art results.